Cognition unveiled Devin Fusion, a multi-model harness that cuts costs by 35% while maintaining frontier-level performance.
Cursor has launched an iOS app, enabling users to control cloud and local agents and manage PRs on mobile.
Vercel Functions now supports deploying HTTP servers via a Dockerfile or Containerfile.
With Vercel Services, you can integrate and manage frontends and backends built on multiple frameworks within a single project.
Hugging Face has integrated EEE and Community Evals to address the fragmentation problem in model evaluation results.
Vercel has expanded audit log events to over 400 and now offers data export via Vercel Drains.
A new training technique is proposed that interleaves the management of 3 layer states to improve GPU memory efficiency.
The South Korean government and major companies are pushing forward a $1 trillion mega-project to secure global leadership in AI infrastructure and the robotics industry.
Microsoft has unveiled Memora, a memory framework that separates stored content from retrieval methods to enhance AI agents' ability to perform long-term tasks.
This explains three approaches and architectural decision factors for efficiently integrating and scaling ElevenLabs' Text to Speech API.
Introduces a technique that uses asymmetric quantization to dramatically lower the storage cost of Late Interaction models while preserving retrieval quality.
A Pull Request to support the DeepSeek V4 model has been merged into the llama.cpp project.
vLLM Semantic Router has unveiled a micro-agent runtime that makes multiple models collaborate at the model API layer.
DiScoFormer was released, estimating a data's Density and Score in a single computation without retraining.
Gemini integrates with Google services like Google Photos to support personalized image generation that reflects users' tastes and style.
DeepSeek is introducing a peak-valley pricing system starting mid-July that charges 2x the standard price during certain hours.
Anthropic's Claude models are now generally available on Microsoft Azure powered by NVIDIA GB300 Blackwell Ultra GPUs.
DeepReinforce released Ornith-1.0, a coding-specialized model built on Gemma 4 and Qwen 3.5.
Deep Agents has launched Dynamic Subagents, which build complex agent pipelines through code-based programmatic orchestration.
BigQuery's new AI.AGG function allows you to summarize and analyze millions of rows of unstructured data using natural language.
Google delivers high-performance AI services through an integrated AI full-stack strategy spanning infrastructure, models, and interfaces.
Amazon is introducing automation and AI into its HR division, reducing human resources and shifting processes toward an app-centric model.
Firefly Aerospace will implement in-orbit AI data processing using NVIDIA Jetson on its next-generation lunar exploration mission.
Claude Code's system prompt has been updated regarding agent security monitoring and tool-use metadata.
AI Briefing